CVE-2026-27893
vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.18.0, two model implementation files hardcode `trust_remote_code=True` when loading sub-components, bypassing the user's explicit `--trust-remote-code=False` security opt-out. This enables remote code execution via malicious model repositories even when the user has explicitly disabled remote code trust. Version 0.18.0 patches the issue.
- Affected products
- Hugging Face, Vllm
- Vllm
- < 0.18.0
- Fix
- Available
- CVSS 3.1
- 8.8 HIGH
- EPSS
- 1.3% (69th percentile)
- Weakness
- CWE-693, CWE-501
- NVD status
- Modified
- Published
- 2026-03-26
CVE-2026-27893 at NVD
No indexed exploits for CVE-2026-27893 yet
Our index is partial: it proves presence, never absence
No exploit for CVE-2026-27893 has been indexed yet. Our index is built from live traffic and upstream syncs, so this page can only say what it knows — not that no exploit exists.